{"ID":6023333,"CreatedAt":"2026-07-08T01:00:23.257252134Z","UpdatedAt":"2026-07-10T01:44:12.350457273Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2607.05748","arxiv_id":"2607.05748","title":"Two Sides of the Same Coin: Learning the Backdoor to Remove the Backdoor","abstract":"The community has recently developed various training-time defenses to counter neural backdoors introduced through data poisoning. In light of the observation that a model learns poisonous samples responsible for the backdoor easier than benign samples, these approaches either use a fixed threshold of the training loss for splitting or iteratively learn a reference model as an oracle for identifying benign samples. In particular, the latter has proven effective for anti-backdoor learning. Our method, HARVEY, leverages a similar yet crucially different technique: learning an oracle for poisonous rather than benign samples. Learning a backdoored reference model is significantly easier than learning a reference model on benign data. Consequently, we can identify poisonous samples much more accurately than related work identifies benign samples. This crucial difference enables near-perfect backdoor removal as we demonstrate in our evaluation. HARVEY substantially outperforms related approaches across attack types, datasets, and architectures, lowering the attack success rate to the very minimum at a negligible loss in natural accuracy. The figure below shows an overview of our methods working principle.","short_abstract":"The community has recently developed various training-time defenses to counter neural backdoors introduced through data poisoning. In light of the observation that a model learns poisonous samples responsible for the backdoor easier than benign samples, these approaches either use a fixed threshold of the training loss...","url_abs":"https://arxiv.org/abs/2607.05748","url_pdf":"https://arxiv.org/pdf/2607.05748v1","authors":"[\"Qi Zhao\",\"Christian Wressnegger\"]","published":"2026-07-07T02:11:08Z","proceeding":"cs.LG","tasks":"[\"cs.LG\"]","methods":"[]","has_code":false}